Journal of Plant Growth Regulation

, Volume 35, Issue 3, pp 744–754

Comparative Proteomic Analysis Reveals Nitrogen Fertilizer Increases Spikelet Number per Panicle in Rice by Repressing Protein Degradation and 14-3-3 Proteins

  • Chengqiang Ding
  • Yan Wang
  • Zhongyuan Chang
  • Siliang You
  • Zhenghui Liu
  • Shaohua Wang
  • Yanfeng Ding
Article

DOI: 10.1007/s00344-016-9579-0

Cite this article as:
Ding, C., Wang, Y., Chang, Z. et al. J Plant Growth Regul (2016) 35: 744. doi:10.1007/s00344-016-9579-0

Abstract

The spikelet number per panicle is established in the early stages of panicle development. Nitrogen fertilizer application before panicle initiation is known to increase spikelet number, which is one of the most important traits in rice productivity determination. However, the basic proteomic mechanism remains poorly understood. The present study shows that nitrogen fertilizer significantly increased spikelet number and grain yield in rice. Proteomic variations were further analyzed in young panicles at the secondary panicle branch initiation and spikelet meristem initiation under nitrogen fertilizer treatment. Proteomic analysis identified 63 proteins with significant differential accumulation in young panicles under nitrogen fertilizer treatment. Proteolysis represents the largest functional category, which suggests that protein degradation is an important pathway in the response to nitrogen fertilizer. Importantly, nitrogen fertilizer significantly reduced 14-3-3 proteins, which interact with key enzymes associated with carbon and nitrogen metabolism, and the rice FT homologue Hd3a. Real-time PCR revealed that Hd3a signaling is also repressed by nitrogen fertilizer in leaves. This study contributes to a better understanding of the regulation of nitrogen fertilizers in the flowering pathway leading to panicle development. The identification of novel genes provides new insight into the profound impacts of nitrogen fertilizer on panicle development in rice.

Keywords

14-3-3 proteins, nitrogen fertilizer Rice Spikelet number Protein degradation 

Introduction

Rice is cultivated in more than 100 countries on five continents, and is the staple food of at least half of the world’s population (Muthayya and others 2014). Growing populations demand higher rice yields, which is limited by flower number. Spikelet number is one of the most important traits in rice productivity determination, and the number of spikelets per panicle is established in the early stages of panicle development (Furutani and others 2006; Li and others 2011). Therefore, increasing the spikelet number per unit area via breeding and/or crop management is vital to the increase of rice yields (Yoshida and others 2006).

Nitrogen is one of the most yield-limiting nutrients in crop production; therefore, its proper management is essential to improving grain yield (Chen and others 2014). From 1977 to 2005, the average grain production per unit area increased 98 %, but synthetic nitrogen fertilizer application increased 271 % over the same period (Ju and others 2009). A vast amount of nitrogen fertilizer is applied to rice every year in China to obtain high yield. Most farmers in China typically utilize the following three nitrogen fertilizer applications during the growing season: before transplantation, at the start of, and upon panicle initiation. The application of nitrogen fertilizers prior to the panicle initiation stage is known to affect inflorescence development, and it increases the flower number per panicle (Yoshida and others 2006). Cytokinin biosynthesis was up-regulated in response to exogenously applied nitrate and ammonium in rice (Kamada-Nobusada and others 2013). Nitrogen is a key limiting factor of agricultural productivity, and it is well known that nitrogen fertilizers regulate rice panicle architecture. The grain number is the most important factor for increasing grain yield under field production conditions, and it is linearly correlated with total plant nitrogen content (Makino 2011). In our previous study, we tested the effects of nitrogen fertilizer on panicle size, and the panicle size of high nitrogen-fertilized plants was significantly larger than that of normal nitrogen-fertilized (control) plants (Ding and others 2014). We also demonstrated that nitrogen fertilizer increases the flower number per panicle by enhancing local cytokinin synthesis in the panicle (Ding and others 2014), but exactly how nitrogen fertilizer regulates the function of genes involved in inflorescence development remains unclear.

The application of “omics” research technologies offers the potential to identify the genes involved in various aspects of nitrogen regulation. Transcriptomic analyses have been utilized to comprehensively view gene expression under nitrogen treatments (Lian and others 2006; Wang and others 2014); however, proteomic approaches can provide additional information on gene regulation. Two-dimensional gel electrophoresis (2-DE) has been widely used to study nitrogen effects on rice growth and development (Ding and others 2011; Hakeem and others 2012). Twelve protein spots were successfully identified via mass spectrometry, which provides new insight into the mechanisms involved in rice root response to low nitrogen stress (Ding and others 2011). These proteins were categorized into classes related to the tricarboxylic acid cycle, adenylate metabolism, and phenylpropanoid metabolism. Furthermore, 47 proteins were identified with significant differential accumulation in young ears under different nitrogen treatments (Liao and others 2012). In the study, Liao and others (2012) found that nitrogen deficiency or oversupply mediated ear growth via differential regulation of hormonal metabolism. This was the first proteomic study to reveal the role of nitrogen in inflorescence development. In another study, the authors examined spike development inhibition in bread wheat, and provided the theoretical basis for studying floral development and the transition during the reproductive growth period (Zheng and others 2013). Their results suggest that retinoblastoma-associated protein FVE, which modulates flowering, might play an important role in wheat spikes development.

According to Furutani and others (2006), the early stage of panicle development can be divided into five sub-stages: (1) early stage of primary panicle branch initiation, (2) late stage of primary panicle branch initiation, (3) secondary panicle branch initiation, (4) spikelet meristem initiation, and (5) floral organ initiation. Here, we conducted proteomic analyses that focused on the secondary panicle branch initiation stage and spikelet meristem initiation stage, because the basic architecture of the panicle is established at these stages. Thus, we studied proteome response to nitrogen regulation, and demonstrated that nitrogen fertilizer treatment increases the flower number per panicle by repressing the flowering pathway, perhaps through modulation of protein degradation and 14-3-3-like proteins in young panicles.

Materials and Methods

Plant Material and Growth Conditions

The experiment was conducted at a Nanjing Agriculture University farm (Jiangsu Province, China; 118.8°E, 32.0°N) during the rice-growing season (May–October, 2009). The soil type was yellow loam soil, and the main physicochemical properties were as follows: pH 6.75, organic matter 13.15 g kg−1, total nitrogen 0.91 g kg−1, rapidly available nitrogen 76.40 mg kg−1, available phosphorus 11.9 mg kg−1, and K 84.9 mg kg−1. Rice (Oryza sativa L. spp. Indica var. 9311) seeds were soaked, pre-germinated for 72 h, and sown in soil. As described in detail previously (Ding and others 2014), 1.73 g N per pot (corresponding to 120 kg N per hectare) was applied to the soil as liquid urea 15 days before the panicle initiation stage in the nitrogen fertilizer treatment, and no additional nitrogen was applied in the control treatment.

As described in detail previously (Ding and others 2014), the panicles were cut off and mixed together when the panicle’s stage was spikelet meristem and floral organ initiation. Panicles were immediately frozen in liquid nitrogen and stored at −80 °C until use. These panicles were used as one repetition. Three repetitions of panicles were prepared for further analysis. The same method was used to collect the panicles of control treatment plants.

Protein Extraction

Protein for 2-DE analysis was performed as previously reported (Ding and others 2011). The total protein concentration was assessed using a 2-DE Quant Kit (GE Healthcare), and BSA was used as a standard.

Two-Dimensional Gel Electrophoresis

Isoelectric focusing (IEF) was performed using a PROTEAN IEF apparatus (pH 3–6, and pH 5–8, Bio-Rad). Two kinds of strips (17 cm, pH 3–6 and 17 cm, pH 5–8) have been used here, and provide optimal overlaps. Thus, the proteins will separate in longer distances with a higher sample loading capacity of gels making protein identification easier. IEF and SDS-PAGE were performed as previously reported (Ding and others 2011). Analytical gels were stained with silver nitrate (Yan and others 2000). Three repetitions of the material preparation and 2-DE were performed to ensure the reliability of the results (Fig. S1).

Gel Analysis

Gel analysis was described in detail previously in Ding and others (2011), (2012). The stained gels were scanned using a GS-800™ Calibrated Densitometer (Bio-Rad), followed by analysis of the protein spots with PDQUEST 8.0 software (Bio-Rad). The spots were automatically detected by the software, and then inspected visually. Spot filtering and editing were examined manually. All artifacts should be removed to confirm spots that split correctly.

In-Gel Digestion and Protein Identification

Protein spots were manually excised from the gels, and subsequently in-gel digested by trypsin according to Chi and others (2010). Peptide mass spectra were obtained on an Applied Biosystem Sciex 4800 MALDI-TOF/TOF mass spectrometer. Data were acquired in a positive MS reflector using a CalMix5 standard to calibrate the instrument (ABI4700 Calibration Mixture). Mass spectra were obtained from each sample spot by accumulation of 900 laser shots in an 800–3500 mass range. For MS/MS spectra, the five most abundant precursor ions per sample were selected for subsequent fragmentation and 1000–2000 laser shots were accumulated per precursor ion. The criterion for precursor selection was a minimum S/N of 50. Both the MS and MS/MS data were interpretated and processed by using the GPS Explorer software (V3.6, Applied Biosystems), then the obtained MS and MS/MS spectra per spot were combined and submitted to the MASCOT search engine (V2.1, Matrix Science, London, U.K.) by GPS Explorer software and searched with the following parameters: NCBInr database, taxonomy of Oryza sativa, trypsin of the digestion enzyme, one missed cleavage site, partial modification of cysteine (carboamidomethylated) and methionine (oxidized), no fixed modifications, MS tolerance of 50 ppm, and MS/MS tolerance of 0.25 Da. Known contaminant ions (keratin) were excluded. A total of 50,346 sequences were searched. MASCOT protein scores (based on combined MS and MS/MS spectra) greater than 61 were considered statistically significant (p < 0.05), and the individual MS/MS spectrum with a statistically significant (confidence interval >95 %) best ion score (based on MS/MS spectra) was also accepted.

RNA Extraction and cDNA Synthesis

Separated panicles and leaves were immediately frozen in liquid nitrogen and stored at −80 °C until use. Total RNA was extracted from the frozen samples using an EZNATM Plant RNA Kit (OMEGA), and was then treated with RNase-free DNase I (OMEGA) to remove any remaining genomic DNA. Isolated RNA was quantified based on the A260 value, and RNA purity was determined from the 28S:18S rRNA ratio on a MOPS/formaldehyde gel. First-strand cDNA was synthesized from 500 ng of total RNA using a PrimeScript RT Reagent Kit (TaKaRa, Dalian, China) and following the manufacturer’s protocol. Three repetitions were performed for each treatment.

Real-Time Reverse Transcriptase Polymerase Chain Reaction (qRT-PCR) Assay

As described in detail previously (Ding and others 2014), real-time reverse transcriptase polymerase chain reactions (qRT-PCR) were performed using an Applied Biosystems 7300 Sequence Detection System and a SYBR Premix Ex Taq Kit (TaKaRa). The PCR consisted of 12.5 μL of SYBR Premix Ex Taq, 0.5 μL of each 10 µM forward and reverse primer, 9.5 μL of water, and 2 μL of template cDNA in a total volume of 25 μL. Cycling was performed using the default conditions of the ABI 7300 SDS Software 1.3.1 using the following cycling conditions: 90 °C for 30 s followed by 40 cycles of 95 °C for 5 s and 60 °C for 31 s. Expression data were normalized using the β-actin expression level (Rieu and Powers 2009). All primer sequences used are listed in Table S1.

Statistical Analysis

In-gel analysis, spots were detected and quantified on the basis of their volume, and the spot volumes were analyzed by Student’s t test between the a nitrogen treatment and control. The threshold for significance was set at p < 0.05 and >2-fold or <0.5-fold (treatment/control). In qRT-PCR analysis, three biological and technical replicates were performed per treatment, and the data presented are the mean values of the three biological replicates, which each had three technical replicates.

Results

Proteomic Changes in Response to Nitrogen Treatment

Here, the proteome of panicle response to nitrogen fertilizer treatment was analyzed at later stages. We did not study the proteome at earlier stages because the panicle is too small and light, which makes the extraction of sufficient protein quantities for 2-DE difficult. To investigate the changes in the rice panicle proteins response to nitrogen fertilizer, samples were taken 15 days after nitrogen fertilizer treatment. To dissect proteomic variation in young panicles caused by nitrogen treatments, proteins were separated by 2D electrophoresis, and then the nature of differentially accumulated proteins was determined with a matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF/TOF MS) analysis. About 3500–4200 protein spots were visualized using silver staining (Fig. 1). The molecular weights of these proteins ranged from 15 kDa to >120 kDa, with a pI range from 3 to 6 (Fig. 1a, b) and 5 to 8 (Fig. 1c, d). Approximately 63 protein spots showed significant differential accumulation in young panicles between treatments. In comparison to control panicles, nitrogen-fertilized panicles had eight up-regulated proteins, and 55 proteins were down-regulated (Fig. 2).
Fig. 1

Analysis of proteins extracted from rice panicles by 2-DE. The first dimension was performed using 17 cm pH 3–6 IPG strips to separate proteins of control (a) and nitrogen fertilizer (N+) (b), and also separated control (c) and nitrogen fertilizer (N+) (d) proteins using pH 5–8 IPG strips. The second dimension was performed using 10 % SDS-PAGE gels. The gel was stained with silver nitrate, and images were analyzed with PDQUEST 8.0 software (Bio-Rad). Arrows indicate the differentially expressed proteins after nitrogen application

Fig. 2

2-DE gels showing the changed expression of proteins affected by nitrogen fertilizer (N+) versus control. The spot numbers are the same as those specified in Table 1, 2 and Fig. 1. Arrows indicate the differentially expressed proteins after nitrogen application

Functional Classification of Identified Proteins

Sixty-three differentially expressed proteins were unambiguously identified using MALDI-TOF MS analyses and NCBI database searching (Tables 1, 2, S2). Fifty-four spots matched known proteins that were annotated according to the Uniprot database (based on BLASTX scores), and were categorized into the following 11 functional groups: proteolysis (20.4 %), metabolism (14.8 %), energy (11.1 %), stress responses (9.3 %), cell structure (9.3 %), transporters (9.3 %), signal transduction (5.6 %), protein synthesis (5.6 %), phenylpropanoids/phenolics (5.6 %), mRNA processing (5.6 %), and unclear classification (3.7 %) (Fig. 3). When compared to control, the proteins showed significant changes with mean ratios above 2.0 or below 0.5 under high nitrogen fertilizer treatment, and the detailed fold changes are listed in Table 1.
Table 1

Down-regulation proteins identified by MALDI-TOF/TOF

Spot ID

Homologue

Accession

Ratio

MW (kDa)/pI

NMP

SC (%)

MASCOT score

Experimental

Theoretical

Signal transduction

 3210

EF hand family protein

Q2RAR8

0.01

32.9/6.1

32.2/5.9

4

9

107

 3307

Putative WD-40 repeat protein

Q69X61

0.03

35.4/5.9

37.6/5.4

8

20

311

 6202

EF hand family protein

Q2QY10

0.02

31.9/6.6

30.6/6.6

12

23

399

Transporters

 6006

GTP-binding protein

Q9SDK4

0.03

19.5/6.6

21.9/6.3

4

22

87

 7006

Putative transmembrane protein

Q7XI54

0.4

19.7/6.9

25.8/7.8

9

23

343

 8006

GTP-binding nuclear protein Ran-1

A2WSI7

0.4

25/7.0

25/6.4

18

48

561

 9213

Putative H+-exporting ATPase

Q8SA35

0.35

33.4/7.7

26.6/6.9

20

64

238

Stress responses

 3008

14-3-3-like protein

A6N0I0

0.07

16.8/6.0

24.9/4.9

1

7

105

 3103

14-3-3-like protein GF14-F

Q06967

0.17

30.3/4.8

29.1/4.8

22

60

269

 7101

Osr40c1

Q8H7M3

0.03

28.8/6.7

38.8/6.3

21

47

816

 8004

Putative glutathione transferase III(b)

Q8LR62

0.34

25.5/7.0

24.1/6.2

12

30

276

 9101

Heat shock cognate 70 kDa protein

Q2QZ41

0.01

29.7/7.3

67.3/5.0

36

33

918

Cell structure

 2206

Alpha-tubulin

Q0PVB1

0.02

34.7/5.7

49.7/4.8

21

45

795

 2910

Putative chloroplast inner envelope protein

Q9FWV2

0.04

112.4/5.7

107.8/5.4

12

9

304

 2914

Chloroplast inner envelope protein

Q7XD45

0.07

112.3/5.7

110.7/5.5

38

37

443

 4904

Villin-3, putative

Q10L71

0.4

120.8/6.2

106.1/5.6

28

25

723

 5903

Villin-3

Q10L72

0.3

120.4/6.3

106.7/5.6

22

23

656

Proteolysis

 2806

Unnamed protein product

B7EA73

0.05

97.4/5.7

99.5/5.4

9

14

144

 2810

Putative aminopeptidase N

Q6ZBX8

0.41

96.5/5.7

98/5.4

62

64

1010

 2904

Ubiquitin-activating enzyme E1 2

B9GBE8

0.35

120.5/5.5

117.1/5.2

15

8

611

 2906

Ubiquitin-activating enzyme E1 2

B9GBE8

0.27

127.2/5.6

117.1/5.2

11

8

627

 2911

Ubiquitin-activating enzyme E1 2

B9GBE8

0.4

126.9/5.7

117.1/5.2

5

3

170

 2918

Ubiquitin-activating enzyme E1 2

B9GBE8

0.50

120/5.6

117.1/5.2

37

47

701

 5807

Prolyl oligopeptidase family protein

Q338C1

0.43

94.0/6.4

84.9/5.8

13

20

402

 7002

Os04g0428900

Q0JD68

0.06

18.5/6.6

16.1/5.3

7

25

240

 7802

Putative aminopeptidase N

Q6ZBX8

0.25

98.0/5.4

97.8/5.4

17

22

80

 8301

26S proteasome regulatory particle non-ATPase subunit8

Q8W424

0.34

37.5/6.9

34.9/6.3

23

55

670

Metabolism

 2907

Carbamoyl-phosphate synthase large chain

B9EXM2

0.21

137.02/5.6

127.7/5.7

47

43

787

 5410

Putative stearoyl-Acyl-carrier protein desaturase

Q8S059

0.42

39.4/6.4

44.9/6.4

28

51

536

 5804

Hypothetical protein OsI_05634

B9F3P2

0.38

98.0/6.4

102.0/5.7

7

10

185

 6204

Spermidine synthase 1

Q9SMB1

0.28

34.5/5.2

35.1/5.2

10

31

100

 6901

Hypothetical protein OsJ-26310

B9F3P2

0.05

108.2/6.5

103.8/6.4

7

12

290

 7012

Allene oxide cyclase

Q8L6H4

0.06

20.5/6.8

26.0/9.3

23

45

593

 7805

5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase

Q2QLY4

0.4

80.2/6.7

84.6/5.9

39

46

1060

Energy

 3306

Putative enolase

Q42971

0.04

35.9/5.9

45.9/5.2

2

7

76

 4306

glyceraldehyde-3-phosphate dehydrogenase

Q7FAH2

0.02

35.4/6.3

36.8/6.3

9

28

501

 5404

Enolase

Q42971

0.02

38.7/6.4

47.9/5.4

9

24

642

 5802

Putative aconitate hydratase

Q6YZX6

0.03

94.9/6.3

98.0/5.7

38

38

568

 6103

Glyceraldehyde-3-phosphate dehydrogenase

Q0J8A4

0.25

30.6/6.6

36.5/6.6

10

33

368

 7504

6-phosphogluconate dehydrogenase

Q7Y248

0.01

49.8/6.7

51.6/6.6

34

63

667

Phenylpropanoids/phenolics

 7211

Putative cinnamyl alcohol dehydrogenase

Q5QM39

0.25

34.5/6.9

36.9/6.3

9

31

219

 8707

Aldehyde dehydrogenase

Q9FPK6

0.25

57.9/7.1

54.5/6.7

31

65

575

Protein synthesis

 4804

H0209H04.1

Q01L37

0.38

74.8/6.3

74.7/5.8

28

37

689

 6003

40S ribosomal protein S12

Q6ZLP8

0.01

13.1/5.3

14.8/5.3

8

60

99

 8603

EF-1-gamma 2

Q6YW46

0.23

52.4/7.0

46.5/6.3

8

24

646

mRNA processing

 8508

Hypothetical protein OsI_18438

B8AXX6

0.05

48.2/7.3

50.9/7.6

32

58

558

 8608

Os05g0437300

Q0DHV7

0.5

52.4/7.2

43.4/6.8

13

32

370

 8609

Os01g0867800

Q0JHE8

0.01

52.3/7.2

49.3/6.5

24

53

942

 1012

No good matched

0.03

     

 2106

No good matched

0.11

     

 2208

No good matched

0.33

     

 5810

No good matched

0.06

     

 6001

No good matched

0.01

     

 6205

No good matched

0.19

     

 7001

No good matched

0.44

     

Significance threshold p < 0.05. The total of peptides are listed in Table S2

Spot ID spots are named accordingly with Figs. 3 and 4, Accession the code of the identified protein in Uniprot database, Ratio defined as the %vol of individual spot of nitrogen fertilizer treatment to that of the corresponding spot at control, MW (kDa)/pI molecular weight and isoelectric point of identified protein, NMP number of peptides matched, SC sequence coverage, Score score of Mascot Search Results

Table 2

Up-regulation proteins identified by MALDI-TOF/TOF

Spot ID

Homologue

Accession

Ratio

MW (kDa)/pI

NMP

SC (%)

MASCOT score

Experimental

Theoretical

Unclear classification

 203

Os01g0143300

Q9FU80

2.23

33.6/4.3

29.4/5.5

4

6

72

 8505

Os07g0295400

Q6YVE2

2.28

46.2/7.1

43.4/6.9

20

43

682

Transporters

 1101

Putative nascent polypeptide associated complex alpha chain

Q75K80

2.19

29.3/4.4

13.7/4.9

5

44

66

Proteolysis

 3002

Thioredoxin family Trp26-like protein

Q5JKQ4

2.15

25.6/4.8

22.6/5.0

6

29

88

Metabolism

 6008

Inosine triphosphate pyrophosphatase

Q7XDP2

2.09

21.7/6.6

21.9/6.3

14

43

634

Phenylpropanoids/phenolics

 4307

Putative isoflavone reductase

Q9FTN5

57.61

35.1/6.2

33.5/5.7

9

25

331

 2207

No good matched

2.13

     

 3312

No good matched

2.07

     

Significance threshold p < 0.05. The total of peptides are listed in Table S2

Spot ID spots are named accordingly with Figs. 3 and 4, Accession the code of the identified protein in Uniprot database, Ratio defined as the %vol of individual spot of nitrogen fertilizer treatment to that of the corresponding spot at control, MW (kDa)/pI molecular weight and isoelectric point of identified protein, NMP number of peptides matched. SC sequence coverage, Score score of Mascot Search Results

Fig. 3

The classification of the differentially changed proteins in response to nitrogen in rice panicles

Proteins Showing Changes in Rice

The results indicate that proteolysis is the largest class compared to the other differentially expressed proteins. In the study, 20.4 % of all identified proteins were involved in protein degradation, and these proteins were decreased by nitrogen fertilizer, with the exception of spot 3002 (thioredoxin family Trp26-like protein). The result was quite different from most previously conducted proteomic studies that analyzed the rice response to nitrogen treatments (Ding and others 2011; Ding and others 2012; Liao and others 2012). We also studied the proteome response to nitrogen, and used the same experimental procedures as those in a previous study. In a previous study (Ding and others 2012), only 10.71 % of proteins were involved in protein degradation. Spots 2904, 2906, 2911, and 2918 were identified as the protein ubiquitin-activating enzyme E1 2 (B9GBE8), and all were down-regulated by nitrogen (Table 1). It was difficult to determine if the four spots were the same protein, because they might contain proteins translated from the same gene with modifications made by ubiquitin. Moreover, the proteins may belong to the same family, which includes three E1 enzymes in rice; therefore, each spot was counted separately. Another protein involved in protein degradation was identified as 26S proteasome regulatory particle non-ATPase subunit 8 (RPN8), which was also down-regulated by nitrogen (Table 1). In addition, the expression levels of E1 (B9GBE8) were tested using qRT-PCR, and the results indicated that nitrogen fertilizer reduced the expression levels of E1 in young panicles (Fig. S2).

14-3-3 proteins play important roles as regulators of carbon and nitrogen metabolism (Yasuda and others 2014). Two protein spots, 3008 and 3103 (Fig. 2; Table 1), were identified as 14-3-3-like proteins, and they were down-regulated by nitrogen fertilizer in rice panicles. The expression of the 14-3-3-like protein GF14-C was also decreased in panicles according to qRT-PCR results (Fig. S2). We also used qRT-PCR to characterize the expression patterns of an additional four genes encoding differentially expressed proteins that were identified by 2-DE (Fig. S2). The mRNA levels of two genes were consistent with the protein levels (Spots 5903 and 8004). These results suggest that nitrogen fertilizer regulates genes at both the transcriptional and posttranslational levels.

Discussion

Ubiquitin–Proteasome System

The ubiquitin–proteasome system plays a fundamental role in the growth and development in plants (Shabek and Zheng 2014). Ubiquitin conjugation is a multistep reaction, sequentially involving three enzymes referred to as ubiquitin-activating (E1), ubiquitin-conjugating (E2), and ubiquitin ligase (E3) enzymes. Recent investigations have shown that the ubiquitin/proteasome degradation system is important for the plant defense response (Takai and others 2002), and E1 enzymes catalyze the first step in this system (Shabek and Zheng 2014). Ubiquitin genes are increased during heat stress in tobacco, potato, and maize (Lyzenga and Stone 2012). In addition, over-expression of ubiquitin gene enhances tolerance to multiple stresses (Guo and others 2008; Tian and others 2014; Zhang and others 2012). Moreover, studies reveal that the system plays an essential role in the adaptation to carbon and nitrogen availability in plants (Sato and others 2011a). NLA encodes a RING-HCa-type ubiquitin ligase, and has an SPX domain (Peng and others 2007). As described in detail previously (Sato and others 2011b), the nla plants displayed an early senescence phenotype in limited nitrogen conditions, but this did not occur when the mutants were grown with sufficient nitrogen. Moreover, E3 interacts with a target protein and adds ubiquitin molecules derived from E1 and E2. Here, we find that E1 enzymes are down-regulated by nitrogen, and the results indicate that the function of the NLA protein is also affected by nitrogen. It is important to note that E1 enzymes can determine ubiquitin ligase activity. ATL31 is a member gene of the ubiquitin ligase family, which is specific in plants. The protein is important for the regulation of the C/N-nutrient response in plants (Yasuda and others 2014). Therefore, the function of the ATL31 protein will also be influenced by nitrogen fertilizer through E1 enzymes.

Another identified protein associated with protein degradation was 26S proteasome regulatory particle non-ATPase subunit 8 (RPN8). In the ArabidopsisRPN10 mutant, plants had decreased sensitivity to cytokinin (Smalle and others 2003); therefore, the results indicated that the 26S proteasome plays an important role in cytokinin signaling. In this study, RPN8 was down-regulated by nitrogen fertilizer, and the expression level of the RPN8 protein may affect the sensitivity to cytokinins. As is known, cytokinin is very important to inflorescence development (Ding and others 2014; Kurakawa and others 2007). In our previous study, another 26S proteasome regulatory particle non-ATPase subunit 10 (RPN10) was also decreased in rice roots by nitrogen (Ding and others 2012). Thus, the results suggest that rice plants could regulate the abundance of 26S proteasome regulatory particle non-ATPase subunit proteins to adapt to the nitrogen level in environment.

Hd3a Signaling

Plant 14-3-3 proteins are mainly thought to be regulators of carbon and nitrogen metabolism, and this is because the proteins have been reported to directly interact with essential enzymes such as the plasma membrane H+-ATPase (Alsterfjord and others 2004), nitrate reductase (Moorhead and others 1996), sucrose phosphate synthase (Bornke 2005), starch synthases (Sehnke and others 2001), and glutamine synthetase (Riedel and others 2001). In addition, proteomic analyses identified 14-3-3 proteins as interactors with ATL31 (Diaz and others 2011; Sato and others 2011a). 14-3-3 proteins also have functions in hormone metabolism and signaling (auxin and brassinosteroid) (Gampala and others 2007; Yao and others 2007). Three 14-3-3-like proteins/genes were down-regulated by nitrogen fertilizer, and the results indicate that both nitrogen and hormone metabolism may be altered by nitrogen fertilizer in young rice panicles via 14-3-3-like proteins.

Heading-date 3a (Hd3a), a rice ortholog of FT, induces flowering in rice (Tamaki and others 2007). Heading-date 1 (Hd1) and Early heading-date 1 (Ehd1) can reduce the number of primary branches in a panicle, resulting in smaller spikelet numbers per panicle, and this occurs independent of the control of flowering time (Komiya and others 2008). Because Hd3a expression is regulated by Hd1 and Ehd1, crop yields in the field may be affected by Hd3a expression in the leaves (Komiya and others 2008). Here, we examined the expression of Hd3a by qRT-PCR, and the results indicate that the gene is down-regulated in leaves by nitrogen (Fig. S3). Furthermore, the results suggest that the flowering pathway is repressed by nitrogen fertilizer in leaves, and that the effects on Hd3a expression are important in controlling panicle development (Fig. 4). The Hd3a protein moves from the leaf to the shoot apical meristem, and 14-3-3 proteins act as intracellular receptors for Hd3a in panicles (Taoka and others 2011). The 14-3-3 protein group has eight isoforms, and four of them (GF14b, GF14c, GF14e, and GF14f) have been reported to interact with Hd3a (Purwestri and others 2009; Taoka and others 2011; Taoka and others 2013). The four isoforms are expressed in all organs, such as leaf and shoots apexes (Purwestri and others 2009). Therefore, a decreased abundance of 14-3-3 proteins will change the signaling of Hd3a in the shoot apical meristem. Thus, we can conclude that Hd3a signaling is repressed by nitrogen fertilizer not only in leaves but also in the panicles (Fig. 4).
Fig. 4

Model explaining the role of nitrogen fertilizer in Hd3a signaling during panicle branching. Effects of nitrogen fertilizer on the expression of Hd3a in the leaves. Real-time PCR was performed in triplicate, and the mean values with SD are shown. The samples were collected at early stages of panicle development. LC leaf under control treatment; LN leaf under high nitrogen fertilizer treatment

In conclusion, the experiments showed that high nitrogen fertilizer increased both spikelet number per panicle and grain yield. Proteomic analysis revealed that nitrogen fertilizer mediated young panicle development via differential regulation of proteolysis, C/N metabolism, stress response, cell structure, and energy in rice panicles. Importantly, high nitrogen fertilizer reduced the levels of proteins in the ubiquitin–26S proteasome system, and also decreased the abundance of 14-3-3-like proteins. The Hd3a-14-3-3 complex has important functions in panicle development according to previous studies. In this study, the Hd3a-14-3-3 complex seems to be vital to nitrogen regulation of rice growth.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Grant Numbers 31401324; 31401324) and the National High Technology Research and Development Program of China (863 Program, Grant Number 2014AA10A605-1).

Author contributions

Chengqiang Ding, Shaohua Wang, and Yanfeng Ding designed research; Chengqiang Ding performed research and analyzed data; and Chengqiang Ding, Yan Wang, and Siliang You wrote the paper.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Supplementary material

344_2016_9579_MOESM1_ESM.pdf (152 kb)
Supplementary material 1 (PDF 152 kb)
344_2016_9579_MOESM2_ESM.pdf (38 kb)
Supplementary material 2 (PDF 38 kb)
344_2016_9579_MOESM3_ESM.pdf (23 kb)
Supplementary material 3 (PDF 23 kb)
344_2016_9579_MOESM4_ESM.pdf (65 kb)
Supplementary material 4 (PDF 64 kb)
344_2016_9579_MOESM5_ESM.xlsx (86 kb)
Supplementary material 5 (XLSX 86 kb) Fragment sequence and the individual score of each protein in this paper

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  • Chengqiang Ding
    • 1
  • Yan Wang
    • 1
  • Zhongyuan Chang
    • 1
  • Siliang You
    • 1
  • Zhenghui Liu
    • 1
  • Shaohua Wang
    • 1
  • Yanfeng Ding
    • 1
  1. 1.College of AgronomyNanjing Agricultural UniversityNanjingPeople’s Republic of China

Personalised recommendations